Learning State Features from Policies to Bias Exploration in Reinforcement Learning.
Bryan SingerManuela M. VelosoPublished in: AAAI/IAAI (1999)
Keyphrases
- reinforcement learning
- learning process
- learning algorithm
- state space
- learning problems
- active exploration
- policy search
- prior knowledge
- supervised learning
- optimal policy
- learning tasks
- feature vectors
- autonomous learning
- machine learning
- function approximation
- action selection
- hierarchical reinforcement learning
- state abstraction
- average reward reinforcement learning
- classification accuracy
- feature selection
- exploration strategy
- learning agents
- state action
- markov decision process
- partially observable
- learning capabilities
- transfer learning
- online learning